Language models improve design of soft robots that move realistically
Bridging Language and Physics: Automated Design of Continuum Robots with Large Language Models
Robotics
Summary
Designing soft robots that bend and move in complex ways is hard because their physical behavior is tricky to predict. The paper shows how language models, like those used for chatbots, can get better at creating robot designs when they get detailed feedback from simulations that show how the robot would really move. The researchers tested this on soft robots with tendons and showed many designed robots actually worked well in simulations and real life. This approach helps computers automatically design soft robots that can perform different tasks.
What this means in practice
- •For robot designers: Automatically generate tendon-driven soft robot designs that are more physically valid and task-capable using simulation-guided language model feedback.
- •For manufacturing engineers: Produce physical prototypes of challenging soft robots whose designs pass simulation checks and demonstrate functionality in real-world tasks.
Authors
Jingyi Chen, Mohan Zhang, Laura Yao, Yingtai Ni, Jianmin Ji, Jie Peng, Song Wang, Tianlong Chen
Abstract
Large language models (LLMs) have recently emerged as a promising tool for automating robot design from high-level specifications, yet they remain ineffective for robots operating under complex physical interactions. This limitation stems from the gap between language-based reasoning and the physical consequences of embodiment, often resulting in designs with low physical validity. In this work, we propose a multi-layered framework, AID-SR, that establishes a closed loop by translating simulator-observed physical states into structured feedback for the LLM designer. Combined with semantic critique, human feedback, and iterative refinement, the framework promotes the generation of physically feasible and functionally meaningful robot designs. We evaluate our approach on tendon-driven continuum robots across a benchmark of 14 tasks spanning reaching, grasping, locomotion, and manipulation. The proposed framework achieves 96.2% rate for passing the simulation feasibility check and by applying a common reinforcement learning training, 26.7% robots can successfully fulfill the corresponding task. We then fabricate three designed robots of AID-SR that successfully complete the task in real-world. These extensive experiments across simulation and real-world environments demonstrate and break the wall of utilizing the LLMs for automated design of continuum robots. The source code and experimental resources are publicly available at https://github.com/UNITES-Lab/AID-SR.